databricks-dolly-15k-pythia-70m-deduped-v1
This model is a fine-tuned version of EleutherAI/pythia-70m-deduped on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 4.3063
Model description
More information needed
Intended uses & limitations
More information needed
Training and evaluation data
More information needed
Training procedure
Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 0.001
- train_batch_size: 8
- eval_batch_size: 8
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 10
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 5.8656 | 1.0 | 1501 | 4.6547 |
| 4.609 | 2.0 | 3002 | 4.4454 |
| 4.1706 | 3.0 | 4503 | 4.3141 |
| 3.8124 | 4.0 | 6004 | 4.3063 |
| 3.4925 | 5.0 | 7505 | 4.3213 |
| 3.1927 | 6.0 | 9006 | 4.3806 |
| 2.9257 | 7.0 | 10507 | 4.4425 |
| 2.6838 | 8.0 | 12008 | 4.5654 |
| 2.4619 | 9.0 | 13509 | 4.7032 |
| 2.2826 | 10.0 | 15010 | 4.8537 |
Framework versions
- Transformers 4.33.3
- Pytorch 2.0.1+cu118
- Datasets 2.14.5
- Tokenizers 0.13.3